67,080 rated or chosen profiles from 2,236 respondents, in up to 15 tasks of 2 profiles varying on 10 attributes.

About this table

Description	2,236 Japanese adults (online quota sample, 2022) rated 15 pairs of hypothetical candidates described by their positions on ten 2021-election policy issues (constitution, counter-strike capability, public works, consumption tax, corporate tax, separate surnames, same-sex marriage, nuclear power, lockdown law, COVID priority). Outcome: likelihood of voting for each candidate, 6-point scale.
Reference	Miwa, H. (2023). 2021 nen sosenkyo ni okeru yukensha no seisaku senko oyobi soten no juyodo to toohyo [Voters' policy preferences, issue importance and vote in the 2021 general election]. Senkyo Kenkyu [Japanese Journal of Electoral Studies], 38(2), 48-62.
DOI	10.14854/jaes.38.2_48
Licence	CC0 1.0
Source data	https://doi.org/10.7910/dvn/jdurc1

Notes

Conjoint table (irw_conjoint_2). Paired candidate profiles (hypothetical candidates' stances on ten policy issues, 2 levels each, Japan 2022 survey), 15 tasks x 2 profiles; attribute order randomized per respondent and recorded. rating = likelihood of voting for each candidate on a 6-point scale from 'definitely would not vote' to 'high probability of voting'; the numeric direction (higher = more likely) is inferred, not labelled in the deposit. No choice column; attribute text is Japanese as stored.

These notes describe the source data, which the IRW reproduces as released. All notes are listed on the Data Notes page.

Design

Topic	candidate choice
A profile represents	person
Country	JP
Language shown	ja
Language of stored levels	ja
Randomization restrictions	unknown
Task order	recorded in the deposit
Profile position	recorded in the deposit

Outcomes:

rating

あなたはそれぞれの候補者について，どのくらい投票する可能性がありますか。(How likely are you to vote for each of the candidates? Judge only from the stated policy positions; six-point scale from 絶対に投票しない 'definitely would not vote' to 高い確率で投票する 'high probability of voting') (translated) [rating, 1 (Definitely would not vote (inferred)) to 6 (High probability of voting (inferred))]

Columns

attr_attackattr_constitutionattr_consumption_taxattr_corporate_taxattr_covid_priorityattr_lockdownattr_nuclearattr_pubworkattr_samesexattr_surnameattrpos_attackattrpos_constitutionattrpos_consumption_taxattrpos_corporate_taxattrpos_covid_priorityattrpos_lockdownattrpos_nuclearattrpos_pubworkattrpos_samesexattrpos_surnamecov_agecov_educationcov_gendercov_party_idcov_prefecture_codecov_vote_pr_2021idprofileratingtask

The layout (one row per respondent, task and profile; choice/rating outcomes; attr_ columns holding each attribute level as displayed) is defined in the IRW conjoint standard. What the outcomes asked is under Design above.

Get the data

Download CSVno account needed Browse on Redivisexplore and query Croissant metadataHugging Face, Kaggle, OpenML

Or load it directly in R or Python (both need a free Redivis account; the first call opens a browser to sign in):

# R
install.packages("remotes")
remotes::install_github("itemresponsewarehouse/Rpkg")
library(irw)
df <- irw_fetch("miwa_2023_policy_positions", source = "conj")

# Python
pip install irw

import irw
df = irw.fetch("miwa_2023_policy_positions", source="conj")

How to cite

@article{Miwa2023,
  author = {Miwa, H.},
  year = {2023},
  title = {2021 nen sosenkyo ni okeru yukensha no seisaku senko oyobi soten no juyodo to toohyo},
  journal = {Senkyo Kenkyu [Japanese Journal of Electoral Studies]},
  volume = {38},
  number = {2},
  pages = {48--62},
  doi = {10.7910/dvn/jdurc1}
}

Respondents	2,236
Rows (one per respondent, task and profile)	67,080
Tasks per respondent (most)	15
Profiles per task	2
Attributes	10

IRW version	v584
Redivis dataset	irw_conjoint_2 v1.4
Redivis dataset DOI	10.57761/hk97-wv30
Manifest pin for this IRW version	v1.4
Metadata source	irw_meta v44.0

miwa_2023_policy_positions